Introduction

The searching for online information by consumers has become a common trend nowadays. Online communication shows rapid growth through the platforms such as social media, websites, blogs, and others, which has raised the interest in the word of mouth as well as electronic word of mouth (Hussain et al. 2017; Yang 2017). One of the long-standing ways of conveying information is through word of mouth (Dellarocas 2003) and it is viewed as one of the most influential factors that affect consumer behaviour (Daugherty and Hoffman 2014). Due to the advancement of online technologies, the new form of word of mouth (WOM) is online WOM communication and is referred to as electronic word of mouth (eWOM) (Yang 2017). The new form of eWOM is different from the traditional WOM as eWOM offers a faster message conveying speed between the users, in which the messages are available online and are retrievable at any time (Huete-Alcocer 2017). The companies can take advantage of eWOM by identifying the factors that motivate their customers to post their opinions and by capturing the impact of their comments (Cantallops and Salvi 2014). On the other hand, eWOM can also be a liability for companies as they are not able to control this factor (Yang 2017). In addition, eWOM has a negative influence on credibility because information communicators and information receivers are kept anonymous (Huete-Alcocer 2017).

In an industry such as tourism, the media that is information and communication technology (ICT) oriented possesses an influence on purchasing behaviour (Cham and Easvaralingam 2012; Cham et al. 2020; Sotiriadis and Van Zyl 2013). Specifically, in the food and beverage industry, Yorganci (2020) found that eWOM has a significant and positive effect on purchase intention. Consequently, this study seeks to investigate the purchase intention of an innovative beverage, bubble tea, which is a modified tea product that has gained immense popularity in recent years across the world (Aji et al. 2020). E-commerce websites enable companies to provide higher efficiency and wider product selection to customers in comparison with the physical retail stores. The purchase nature of Malaysia considering the bubble tea is altering according to the spread of coronavirus during the end of 2019 and the entire 2020. However, as per prediction, the growth of the new bubble tea market will be induced with the revival of the economies across the world. Combined with the rising interest of consumers in this innovative beverage and their eagerness to experiment with different flavours have a significant impact on the growth of the market (Fortune Business Insights 2020). Furthermore, there are more than 100 brands of bubble tea in Malaysia (Aji et al. 2020). Bubble tea market can also be segmented by type of tea such as black tea, green tea, oolong tea and white tea, it can as well be further differentiated by flavours such as original, fruit flavour, chocolate flavour, coffee flavour and others (Fortune Business Insights 2020). Thus, the bubble tea market is particularly competitive. The analysis of purchase intention, especially regarding the newly released flavour of bubble tea, will be critical to the practitioners because of the market growth opportunity and competitive nature of this market. On the other hand, consumers have restricted their movement from home to the physical store. They will rely more on eWOM. The consumer will likely search for information about the new bubble tea through online platforms (Samoggia et al. 2020). The appropriate platform for eWOM is the social media websites. It is noticed that eWOM information has become more reliable and trustworthy due to the reduction in anonymity. The purchase intentions of consumers are influenced by conversations in social media that refer to brands (Samoggia et al. 2020). However, not all eWOM information has the potential in influencing the purchase intention of consumers.

There are various ways through which eWOM information can arise in social media. People can be affected by the information from the communication platform which is computer-mediated through the central route and peripheral route (Erkan and Evan 2016). Information Adoption Model (Sussman and Siegal 2003) articulates the role of information via four components, which comprise argument quality (a proxy for the central route), source credibility (a proxy for the peripheral route), information usefulness and information adoption (Erkan and Evan 2016). The false information shared to the consumer who adapts the previous review will result in the loss of purchase intention. The information in social media is easy to access and the users may be anonymous. This causes the quality and credibility of the information on social media to be lost. Therefore, these two characteristics are critical to explaining the influence of eWOM information.

Moreover, the extent to which users believe that the information given on an online platform fulfils the requirements of their tasks is essential. When the consumers avail good information fit their tasks such as the search about the product like new bubble tea, it will impact their purchase intention. The application of information task-fit enables the visitors to lower specific but common task-oriented information deficit (Hussain et al. 2018). Therefore, it also resolves task-oriented inquiries. For instance, a website that provides complete insight about a product, then it is referred to as having a good information task-fit (Istanbulluoglu 2017). Therefore, the good information task-fit with relation to new bubble tea on the online platform will help to influence the purchase intention of the consumers. Although several previous studies have been carried out to understand the impact of social media eWOM upon the purchase intention of consumers, an in-depth analysis has not been done (Aji et al. 2020; Lim et al. 2019). This study includes the importance of information task-fit to identify its influence on purchase intention of new flavoured bubble tea. Thus, this study fills the gap in the past research through the incorporation of information task-fit in the Information Acceptance Model (IACM).

In addition to the characteristics of eWOM information, consumers’ behaviours towards information in terms of the degree of favourable and unfavourable behaviour is also important to predict the consumers’ decisions. Ajzen (1991) defined it as an attitude, in which a positive attitude leads to a stronger willingness for an individual to perform the behaviour. Besides that, in the process of seeking information from social media, people tend to select useful information and adopt them. As a result, consumers would perceive the information as useful if the information acquired meets their needs. Thus, the need for information is another perspective of consumers’ behaviours to be considered in this study.

Nevertheless, if people perceive information as useful, they will tend to engage with the information (Erkan and Evan 2016). As a result, it is significant to investigate the relationship between information usefulness and the eWOM information characteristics, as well as consumers’ behaviours towards eWOM information. Information usefulness serves as the main predictor for information adoption (Sussman and Siegal 2003) and purchase intention (Lee and Koo 2015). Therefore, this study aims to investigate the purchase intention of new flavoured bubble tea by combining the predictors from characteristics of information and consumers’ behaviours towards information through an indirect path of information usefulness and information adoption. This study contributes to the literature of purchase intention by proposing the role of information task-fit in an Information Acceptance Model in the investigation.

Literature review and hypothesis development

Electronic word of mouth (eWOM) using social media

Drawing from its impact on influencing consumers’ purchase intention, Electronic Word of Mouth (eWOM) on social media websites has drawn attention from various scholars (Iyengar et al. 2009; See-To and Ho 2014; Wallace et al. 2009; Wang et al. 2012; Erkan and Evans 2018). Social media website is widely accepted as an appropriate platform for eWOM by businesses (Canhoto and Clark 2013; Erkan and Evans 2014, 2018; Kim et al. 2014; Toder-Alon et al. 2014). Social media platform gains its popularity among consumers for information seeking (Baird and Parasnis 2011; Naylor et al. 2012; Goodrich and de Mooij 2014; Barreda et al. 2015; Schivinski and Dabrowski. 2016; Erkan and Evans 2018). Peer opinion and experience on the social media website has become one of the most useful information sources for the consumers, as compared to company-generated information (Brown et al. 2007; Mazzarol et al. 2007; Munnukka et al. 2015; Erkan and Evans 2018). Social media website allows Internet users to communicate with their networks, share their thoughts using written text, pictures or videos easily through forwarding post or sharing them through other mobile applications (Kozinets et al. 2010; Chu and Kim 2011; Moran and Muzellec 2017; Erkan and Evan 2016). WOM has always been considered as one of the most influential factors in affecting consumer decision-making. Similarly, the increased number of consumers sharing their thoughts on social media has made eWOM possess the ability to influence consumers’ purchase intentions (Cham et al. 2021; Daugherty and Hoffman 2014).

Social media website has successfully transformed consumers from passive observers to active participants (Riegner 2007; Chu and Kim 2011; Daugherty and Hoffman 2014). EWOM can be easily generated unintentionally, especially when consumers display their preferences to their network interaction with brands that are associated with posts such as liking, commenting, and posting (Alboqami et al. 2015). Advertising information can also be posted and shared on brands official accounts via social media websites (Alboqami et al. 2015). In general, the content on social media websites can be generated by either marketer or consumer, and both can spread rapidly among users (Jansen et al. 2009; Wolny and Mueller 2013; Alboqami et al. 2015). It is noticeable that some content gets spread rapidly and reaches a wider range of consumers, while some received inadequate reactions (Alboqami et al. 2015). This showed that the influence of information defers from one to another person where the same content may evoke different notions among receivers (Chaiken and Eagly 1976; Cheung et al. 2008; Erkan and Evans 2018). Consumers who experience eWOM will comprehensively critique the information for adoption (Erkan and Evans 2016). Therefore, to better examine the intention of consumers’ adoption of information on a social media platform for decision-making in the purchase of new bubble milk tea product, this study adopted the Information Acceptance Model (IACM) developed by Erkan and Evans (2016) with the extension of information task-fit.

Information acceptance model (IACM)

Information Acceptance Model (IACM) by Erkan and Evan (2016) is the extension of the Information Adoption Model (IAM) with the integration of Theory of Reasoned Action (TRA). The Information adoption model (IAM) was developed by Sussman and Siegal (2003) based on the dual-process model of informational influence to understand how people internalised the information they received (Erkan and Evans 2016). IAM successfully narrowed the scope of the research information adoption process by integrating Technology Acceptance Model (TAM) with the Elaboration Likelihood Model (ELM) which suggest that people can be affected by a message in two routes, namely, central, and peripheral (Erkan and Evans 2018). IAM is strongly applicable for eWOM studies as it explained the information on computer-mediated communication platforms (Cheung et al. 2008, 2009; Shu and Scott 2014; Erkan and Evans 2016). Although IAM is the most commonly adopted model in the study, Erkan and Evans (2016) argued that the model focuses on the characteristics of information and the influence of eWOM on consumer purchase intention, depending on the consumers’ behaviour towards eWOM information. The argument was also in line with the review by Knoll (2015) where the review was conducted about advertising on social media. The integration of TRA and IAM formed IACM, it represents the behaviour of consumers towards eWOM information by expanding the notion of information adoption and explains the process of influencing behavioural intention (Erkan and Evans 2016). TRA assumes behaviour intention is decided based on attitude and subjective norm and it is frequently used by researchers in the study of the relationship between eWOM and purchase intention (Prendergast et al. 2010; Cheung and Thadani 2012; Zhang et al. 2014; Reichelt et al. 2014; Erkan and Evan 2016). IACM assumes that consumers who adopted eWOM information are more likely to develop purchase intention.

Information adoption and information usefulness

Information adoption is shown in the process of individual behavioural intention through intentionally adopting information (Cheung et al. 2008; Wang 2016). Social media has increased consumers’ sharing of information, opinion, and experience, which served as a good information source for persons who are seeking and adopting information when the information is found to satisfy the social media user’s need. Cheung and Thadani (2012) found that information adoption being one of the determinants affecting consumers’ purchase intention. Subsequent research by Torres et al. (2018) further identified information adoption on social media tends to affect the purchase intention of an individual and there is a positive relationship between eWOM information adoption and purchase intention. Therefore, this study anticipates a positive relationship between information adoption and purchase intention.

H1

EWOM information adoption is positively related to consumers’ purchase intention.

Information usefulness refers to new information that is informative by providing supporting thoughts to individuals’ perceptions in enhancing their performance (Bailey and Pearson 1983; Davis 1989; Cheung et al. 2008; Yeap et al. 2014). Previous research considered information usefulness being the main predictor of information adoption and purchase intention (Davis 1989; Sussman and Siegal 2003; Erkan and Evans 2016). Chu and Kim (2011) highlighted that social media users who engage in a great amount of eWOM information tend to develop greater intention in information adoption when an adequate amount of information is perceived to be useful (Arumugam 2016; Erkan and Evans 2016). Information adoption based on information usefulness is highlighted by Liu and Zhang (2010) in the case of eWOM where information is obtained from other consumers who have prior experience in purchasing similar products. Research also shows a positive relationship between information usefulness and purchase intention (Cheung 2014; Dachyar and Banjarnahor 2017). Thus, this study hypothesises that information usefulness is related to information adoption and purchase intention.

H2

EWOM information usefulness is positively related to eWOM information adoption.

Information quality

As information can be easily accessed by social media users, information quality and credibility have become more critical for consumers (Reichelt et al. 2014; Yoo et al. 2015; Erkan and Evans 2016). Information quality is defined as the strength of a persuading message in influencing consumers’ purchase intention (Park et al. 2007; Yeap et al. 2014). The concept of information quality is elemental with the dimension of accuracy, completeness, and timeliness in supplying information that meets the expectations of information users (Kahn et al. 2002; Wicklund 2010). Park et al. (2007) discovered a strong positive relationship between information quality and purchase intention; Erkan and Evans (2018) also found a similar relationship. Information quality being influential through eWOM have shown positive relation towards information usefulness, which indirectly impacted the purchase intention (Xue et al. 2018). Therefore, based on IACM, this research predicts the eWOM information quality is positively related to eWOM information usefulness.

H3

EWOM information quality is positively related to eWOM information usefulness.

Information credibility

Information credibility is the initial factor in the persuasion process of an individual and it is defined as a trustworthy source (Wathen and Burkell 2002). Information provided by highly credible sources are seen as valuable and they promote the transfer of knowledge, it is also the initial factor in the individual persuasion process (Erkan and Evans 2016). The credibility of information acts as the main determinant in consumers’ decision-making, and earlier researches have found a positive relationship between information credibility and consumer purchase intention, especially when the information is found to be useful and adaptable (Park et al. 2007; Prendergast et al. 2010; Hui 2017; Torres et al. 2018). Therefore, this research predicts that the credibility of information on eWOM is positively related to information usefulness.

H4

EWOM information credibility is positively related to eWOM information usefulness.

Information task-fit

To achieve a positive impact on consumers’ experience, the information provided by marketers should meet the need of consumers’ needs (Tarkang et al. 2020). Dedeke (2016) highlighted information task-fit being the standard of consumers’ assessing the appropriateness of product information on a website. Later research also found a positive relationship between information task-fit and eWOM impact on purchase intention (Tarkang et al. 2020). When information on social media satisfied the need of the social media users, it is considered as information task-fit, in other words, the degree of informativeness, resourcefulness, and helpfulness of information influence consumers’ assessment of the usefulness of the information and their adoption of the information (Loiacono et al. 2002; Pavlou et al. 2007). Therefore, this study projects information task-fit on eWOM is positively related to information usefulness and indirectly affected the information adoption that influence purchase intention.

H5

EWOM information task-fit is positively related to eWOM information usefulness.

Needs of information

A human being is known to process information and use them systematically on forming behaviour. This theory is proven by Fishbein and Ajzen (1975) in the development of the Theory of Reasoned Action (TRA). Studies have also shown that consumers behaviour in advice seeking and opinion seeking are the motivators for WOM engagement (Sundaram et al. 1998; Hennig-Thurau et al. 2004; Chu and Kim 2011; Wolny and Mueller 2013). Hoffman and Novak (1996) also pointed out that consumers engaged in website interactivity to obtain relevant information tend to decide on online purchases. It is predicted that there is a high likelihood that consumers found information on social media to be useful and further adopting it, thus, this study anticipates the need for information on eWOM is positively related to information usefulness that indirectly stimulates purchase intention.

H6

Needs of eWOM information is positively related to eWOM information usefulness.

Attitude towards information

Consumer attitude is one of the determinants that many researchers had examined with regards to consumers purchase intention and eWOM (Park et al. 2007; Prendergast et al. 2010). The Theory of Reasoned Action (TRA) (Fishbein and Ajzen 1975) reflects the relationship between attitudes and behavioural intention, which later extended to the Theory of Planned Behaviour (TPB) by Ajzen (1991). Attitude is formed through a person’s evaluation of his or her performing behaviour (Sheeran et al. 2002), while evaluation is the cognitive assessment of information. Erkan and Evans (2016) verified the importance of consumer behaviour in determining the influence of eWOM information. Therefore, this study hypothesises that attitude towards eWOM information is related to the usefulness of the information in evaluation, which also influences the consumers’ purchase intention.

H7

Attitude towards eWOM information is positively related to eWOM information usefulness.

Based on all the proposed hypotheses, the following research framework has been derived as shown in Fig. 1.

Fig. 1
figure 1

Research framework

Full size image

Methodology

Purposive sampling technique is used to select the respondents of this study. The respondents of this study are the purchasers of bubble tea drinks and who are also social media users. The minimum sample size of this study is 92. The sample size is calculated using G*Power statistical software, by applying 0.15 medium effect size of f2 and power of 0.80 at 0.05 significance level. A self-administrated online questionnaire was distributed to the respondents in May 2020 for 2 weeks, which was during the lockdown period of Malaysia. 222 sets of questionnaires had been used for data analysis after screening for the questionnaires with straight-lining patterns. Partial least squares-structural equation modelling (PLS-SEM) was performed to analyse data collected using SmartPLS software. The assessments comprise assessment of measurement model and assessment of structural model. Harman single-factor test, which is a statistical remedy, was performed to identify potential common method bias. The total variance explained was 44.36% for a single-factor. The value is lower than the threshold value of 50%, which indicated the absence of common method bias.

The questionnaire comprised three sections. Section A was the questions regarding the respondents’ profile such as gender, age and others. Section B was the questions related to Information Quality, Information Credibility, Needs of Information, Information Task-Fit, Attitude towards Information, Information Usefulness and Information Adoption. The questions of Section C focussed on Purchase Intention. Reliable resources were used to design the measurements of the variables: 4 items from Park et al (2007); and Erkan and Evans (2018) for information quality; 4 items from Prendergast et al (2010); and Erkan and Evans (2016) for information credibility; 4 items from Chu and Kim (2011); Gökerik et al. (2018); and Erkan and Evans (2016) for needs of information; 3 items from Dedeke (2016) for information task-fit; 3 items from Bailey and Pearson (1983); Cheung et al (2008); Erkan and Evans (2018); and Gökerik et al (2018) for information usefulness; 4 items from Cheung et al (2009); and Erkan and Evans (2018) for information adoption; 3 items from Park et al (2007); and Erkan and Evans (2016) for attitude towards information; and 4 items from Coyle and Thorson (2001); and Erkan and Evans (2016) for purchasing intention. Each of the items was measured using a five-point Likert scale, where (1) indicated strongly disagree, while (5) indicated strongly agree.

Results and findings

Based on Table 1, there was 40.5% of male respondents, while 59.5% was female respondents. The majority of the respondents were aged between 20 to 24 years (32.9%). Most of the respondents (35.6%) purchased three to four cups of bubble tea in a month. Most of the respondents spend more than 4 h on social media (43.2%).

Table 1 Demographics of the respondents
Full size table

The assessment of the measurement model includes the checking of convergent validity and discriminant validity. Based on Table 2, the results of the Average Variance Extracted (AVE) showed that all values for the variables are more than 0.5 and thus convergent reliability is achieved. The loadings for all indicators are more than 0.708, except PI4 with loading of 0.698. One item from information task-fit (ITF3) and two items from information adoption (IA3 and IA4) are eliminated from the analysis due to the low loading values. Loading value of 0.5 is acceptable if the value of AVE exceeds 0.5 (Byrne 2016). As a result, indicator reliability is achieved.

Table 2 Convergent validity and reliability
Full size table

Table 3 presents the results of the Heterotrait-Monotrait (HTMT) Ratio of Correlations. All values are more than the threshold value of 0.85 (Kline 2015) and value one does not fall within the range of all the HTMT’s confidence intervals. Thus, discriminant validity is achieved.

Table 3 Heterotrait-Monotrait (HTMT) ratio of correlations
Full size table

The assessment of structural model includes the analysis for path coefficients and the results are presented in Table 4. Information adoption (β = 0.453, p < 0.01) is positively related to purchase intention, while information usefulness (β = 0.731, p < 0.01) has significant positive relationship with information adoption. Thus, H1 and H2 are supported. Information quality (β = 0.119, p < 0.05) and information credibility (β = 0.201, p < 0.01) are positively related to information usefulness. Therefore, H3 and H4 are supported. Information task-fit (β = 0.313, p < 0.01) show positive and significant relationship with information usefulness. As a result, H5 is supported. Needs of Information (β = − 0.079, p > 0.05) is not significantly related to information usefulness and the relationship is negative. Thus, H6 is not supported. Attitude towards information (β = 0.324, p < 0.01) show positive and significant relationship with information usefulness. Consequently, H7 is supported.

Table 4 Path coefficient and hypothesis testing
Full size table

Table 5 summarises the effect size, coefficient of determination and predictive relevance of the model. Effect size (f2) is used to assess the impact of predictors on endogenous variables. Information quality, information credibility, information task-fit and attitude possess a medium effect on information usefulness as the values are between 0.02 to 0.15, while the effect of needs of information on information usefulness is small as the value is less than 0.02. The impact of information usefulness on information adoption is substantial as the value is greater than 0.35. Information adoption shows a medium effect on purchase intention as the value is 0.259. The issues of collinearity were cross-checked via Variance Inflator Factor (VIF). The VIF values for all constructs were less than 5.0 and thus indicated that there were no collinearity issues (Hair et al. 2017). In terms of R2, 60.0% of the variations in information usefulness were explained by the explanatory variables. 53.4% of the variations in information adoption were explained by information usefulness. 20.6% of the variations in purchase intention (such as in Wang and Yang 2008) were explained by information adoption. All predictors exhibit predictive relevance for endogenous as all Q2 values are greater than zero.

Table 5 Summary of effect size, coefficient of determination and predictive relevance
Full size table

The PLSpredict was performed to identify the predictive analysis of the model. The degree of predicting error of the information usefulness, information adoption and purchase intention is presented in Table 6. As all values of \({Q}_{\mathrm{predict}}^{2}\) were greater than zero, then a comparison between the PLS-SEM model and the naïve linear regression (LM) model can be conducted. The root mean square error (RMSE) was used for the comparison. Medium predictive power can be found in the PLS-SEM model as the majority of the RMSE values of the PLS-SEM model were lower than the RMSE values of the LM model.

Table 6 PLSpredict assessment of variables
Full size table

Discussion

This study reveals that information adoption is positively related to purchase intention, which is in line with the finding of Erkan and Evans (2016). Therefore, the users who adopt the information in eWOM via social media are likely to form a purchase intention, in which the information has been exchanged among friends (Erkan and Evans 2018). The finding of the positive significant relationship between information usefulness and information adoption is also consistent with the finding of Erkan and Evans (2016). When the users perceived that the eWOM information is useful, they will tend to engage in the eWOM information.

Information quality has a positive relationship with information usefulness, which is consistent with the findings of Xue et al (2018). As the information via eWOM is achievable by the users, the quality of the information is the concern. Consumers find the information is useful if the information quality satisfies their demands. Information creditability is also positively related to information usefulness. The finding is supported by the study of Park et al (2007). Therefore, the persuasion of consumers relies on credible information provided via eWOM. Transfer of knowledge is enabled if the information is achieved from a credible source and it turns into an initial factor in the individual persuasion process (Erkan and Evans 2016).

Information task-fit is found to have a significant positive relationship with information usefulness. This finding is tallied with the result of Larsen et al (2009) that task-technology fit affects the perceived usefulness of a system. If the eWOM information meets the tasks need of the users, the users will perceive the eWOM information as useful.

However, needs of information show an insignificant relationship with information usefulness. The result is inconsistent with the finding of Erkan and Evans (2016). This is because not all eWOM information is useful as the consumers might only search for the relevant information about the new flavoured bubble tea that determines their purchase intention. Attitude towards information is also positively related to information usefulness. This is consistent with the finding of Wan and Shen (2015) that the relationship between attitude and perceived usefulness is significant. Therefore, the extent to which the users perceived the benefit in using the eWOM information contributes to the perceived usefulness of eWOM information.

Theoretical implications

Theoretically, this study contributes to the literature of purchase intention in the context of an innovative beverage, bubble tea, by incorporating the role of eWOM. The acceleration of online platforms leads to the significant role of eWOM in consumer decision-making behaviour (Cheng et al. 2019; Daowd et al. 2020). Therefore, this study uses Information Acceptance Model to identify the adoption of eWOM information in developing purchasing intention. Information usefulness, which is determined by information quality, information credibility, information task-fit, needs of information, and attitude towards information, is used to predict eWOM information adoption. Information task-fit, which measures the extent to which information on social media satisfied the need of the consumers, is introduced as one of the predictors of information usefulness.

Practical implications

The market of versatile teas with innovative bold flavours offers a wide range of products, which can be differentiated by the types of tea and flavours (Fortune Business Insights 2020). Therefore, planning and strategize based on the development of this market is critical for the practitioners due to the competitive nature of this market. EWOM has become meaningful due to the establishment of cyber technologies, in which the Integrated Marketing Communication (IMC) designed by the marketers rely on the online information adoption behaviour (Hussain et al. 2018). This study reveals that the decision-making process of the purchase of the new flavoured product of the trendy beverage is connected to eWOM information adoption. To enhance the purchase intention of the newly launched flavour of bubble tea, eWOM information is the key consideration for practitioners to enhance sales. The practitioners should enrich their marketing effort to boost the online reviews of their products as the adoption of eWOM information is relied on how consumers perceive the information as useful. As the perspectives of eWOM information such as information quality, information credibility, information task-fit and attitude towards information are the predictors of information usefulness, the practitioners should focus to improve the quality, credibility and relevancy of the online reviews for their new products. Regular monitoring of the online reviews can be used to achieve this objective. Alternatively, businesses or practitioners can create virtual spaces on their websites enabling the consumers to leave comments and share opinions regarding the products and services (Vallejo et al. 2015). The information about the products that can form a positive attitude should also be taken into consideration by the practitioners. As a result, practitioners will gain greater control over their online reviews.

Limitations and future direction

This study focuses on the perception of consumers on the predictors that explain the information usefulness of eWOM. Quantitative variable such as the quantity of information is not included in the investigation of this study. Future research can include the influence of quantitative variables by using different analysis methods such as regression analysis. In addition, cross country analysis also can be conducted to compare the perceptions of consumers in terms of eWOM information for different countries.

Conclusion

This study investigates the influence of various eWOM information characteristics and consumers’ behaviours towards eWOM information on purchase intention via information usefulness and information adoption. Different from the study of Erkan and Evans (2016), this study incorporates information task-fit in the Information Acceptance Model and it is found significant in determining the information usefulness. The role of information task-fit becomes momentous during the period of pandemic as opposed to the finding of Dedeke (2016) who found that information task-fit was not relevant in explaining the purchase intention. The consumers expect the eWOM information to meets their tasks need and, therefore, will perceive the eWOM information as useful. In addition, this study reveals that needs of information do not relate to information usefulness during the period of pandemic, which is inconsistent with the finding of Erkan and Evans (2016) before the period of pandemic when needs of information are still relevant to determine purchase intention. Consumers can easily access bundles of eWOM information, but not all eWOM information is useful as the consumers might only search for the relevant information that shapes their purchase intention. Also, shopping websites are more superior to social media before the pandemic, in which shopping websites are boosted by information quantity, information readiness, detailed information and dedicated information (Erkan and Evans. 2018). The finding of this study discloses the importance of eWOM during the pandemic and its role in predicting purchase intention. With the growing importance of eWOM in a business, understanding the characteristics of eWOM information and the way that consumers’ behaviours contribute to purchase intention is significant to the practitioners from the competitive industry such as the food and beverage industry. This study contributes to the determination of critical predictors of eWOM information in determining the purchase intention during the pandemic.